In this project we examine the relationship between marital income dynamics (wives' higher earnings, in particular) and the mental and physical health of husbands and wives at midlife (51- 61 years old) using a life course of administrative earnings data and longitudinal latent class analyses. This is a demographically critical issue because wives are increasingly likely to be the primary earner. For example, 22% of wives earned more than their husbands in 2007, compared to only 4% in 1970. Further, wives' higher earnings destabilize traditional ideas of men as breadwinners and have consequently been linked with poorer physical and mental health for husbands of wife breadwinners. However, most prior research focuses only on men and relies on short-term snapshots of wives' higher earnings contributions (i.e., one to three year periods), despite recent evidence that wives' higher earnings are unstable and fluctuate widely over the life course. These gaps leave many unresolved questions about the relationship between wives' higher earnings and the health of married men and women. We address these unresolved questions by conducting longitudinal latent class and regression analyses of 30 years (1962-1991) of couple-level Social Security Administration earnings data from one cohort of the Health and Retirement Study (couples born between 1931 and 1941). We link these couple- level life course earnings data to husbands' and wives' physical and mental health in 1992 (when the sample is 51 to 61 years old). Specifically, we propose to: 1) identify life course dynamics of wives' higher earnings (i.e., length, frequency, transitions into/out of, and life course stage of wives' higher earnings) that may relate to mental and physical health at midlife; 2) examine how these dynamics of wives' higher earnings differentially relate to men's and women's health at midlife; and 3) explore how these documented health relationships are confounded by covariates including demographics, education, and low marital income and/or are caused by health selection as measured by childhood health. This research will be the first to identify distinct and varied patterns of wife breadwinning across a near life course of earnings. Linking these latent class patterns of wife breadwinning to midlife health can illuminate an under-studied correlate of poorer health among aging married couples and a potential cause of gendered health differences within marriage, thereby making important contributions to scholarship on the SES gradient in health, income and health in marriage, and gender differences in health. PUBLIC HEALTH RELEVANCE: The proposed project will advance scholarship on the SES gradient in health, on marital income and health, and on gendered health differences for aging men and women. The findings from this research can be used by public health and medical practitioners to develop psychosocial interventions for families adjusting to the changing economics of marriage.